01 September 2009

North Coast Harbor Bridge proposals

The Cleveland Plain Dealer has a story up about proposals for a new opening footbridge at North Coast Harbor, at Cleveland, USA. The headline is North Coast Harbor pedestrian bridge proposals by Miguel Rosales are outstanding.

Readers may remember Rosales as the architect behind an ambitious, pricy proposals for a "wave bridge" at Portland, Oregon. That design eventually came to nothing, with Rosales replaced by a different architect, Donald MacDonald. Essentially, Rosales' preferred option was decided to be way too expensive for a large-span light rail bridge.

His* latest designs comprise six options for a much shorter, lighter structure, a 27m span opening footbridge across a harbour breakwater (pictured, right). There are two double-leaf bascule designs, two bobtail swing bridges, a very unusual single-leaf bascule, and a curling bridge.

(*I'll follow the Plain Dealer in attributing the designs essentially to Rosales, as there isn't much evidence of the contributions of his collaboratoring consulting engineers, Schlaich Bergermann and Wilbur Smith Associates.)

The designs are out for public consultation, with the City of Cleveland looking to narrow the six options down to two. They have a handy report [PDF] adding technical detail to the pretty pictures.

The choice of opening bridge types seems a little odd, but may be driven by site constraints which include a height limit due to a nearby airport. There's no lifting bridge option presented (although they are very difficult to make attractive), and for a footbridge span of 27m, I'd expect a single-leaf bascule to be the obvious choice, as it can be designed to have low operating costs while opening quickly, and they give the possibility of combining low maintenance with interesting aesthetic choices (see the recent Foryd Harbour Bridge design competition for two examples).

Twin bascule leafs are more difficult to operate and maintain, requiring mechanical interlocks to stitch the two leafs together in the closed position. These are vulnerable to dirt and jamming, especially where the geometry will exaggerate the problems of thermal expansion, as is the case on both Rosales' twin-leaf options, which are cranked or curved in plan (pictured, left).

Rosales seems to have made little or no effort to counter-balance either the bascule or the swing bridge options, which adds to operating cost and decreases operating speed. The swing bridge options (pictured, right) may alleviate this somewhat, as they are back-stayed cable-stayed bridges, but only at the cost of failure-prone rotating mast-heads (because the forestays rotate in plan while the back stays don't move).

Even Calatrava's moveable bridge designs (Samuel Beckett, Puente de la Mujer) are better balanced, with proper "bobtails". Not one of the six design options has its centre of mass anywhere near its centre of rotation, increasing machinery costs and also the consequences of failure (an unbalanced opening bridge is far more likely to be damaged if critical equipment fails).

Even more odd is the fact that three of the designs (all the bascules) are drawbridges. In a conventional opening bridge, the deck is moved directly by a pinion or hydraulic ram mechanism - only the lifting bridge is normally moved indirectly, by cables. To reduce fatigue stress in the cables, they pass over large-diameter sheaves, which are hard to hide in an aesthetically pleasing way.

However, in a conventional lifting bridge, the cables and machinery are only stressed while the bridge opens - in the closed position, the deck sits on fixed supports. (The lifting bridge also has the advantage that the deck is generally counterweighted). This would not be true of Rosales' drawbridges, as visually you would expect the back stay (which is also the draw cable) to be required to support loads even while the bridge is closed. This would imply that the mechanical elements are permanently subject to fluctuating load, and the consequences of mechanical failure are all the more significant (not to mention the difficulties of maintaining the equipment). I'm sure there are ways of addressing these issues, but I'd think there are very good reasons why virtually no modern opening bridges are drawbridges.

A more rational layout would dispense with the back-stays entirely and have the bridge operated by hydraulic rams (or a rack-and-pinion mechanism, if counterweights were added).

So far as I know, the layouts shown for the bascule and swing bridges are therefore essentially without precedent, which ought to give the City of Cleveland a few concerns over likely maintenance and operational problems.

The sixth option is however not without precedent: it's to some extent a blown-up version of the Paddington Basin Rolling Bridge, although with simple (and as a consequence less attractive) mechanisms (pictured left, closed, and below right, open). It again has the problem that the mechanical elements are always under fluctuating load, and it would be amazing if it could be done on the scale of North Coast Harbor without regular maintenance problems.
It's also difficult to see quite how it works in the closed position, as it seems to consist of a series of four hinges, with no obvious way to carry any significant bending moment.

I'm not convinced that the bridge selection process adopted by Cleveland will lead to the best solution. The lead designer has discarded every conventional design option before the public are consulted, so their choice is between six challenging, high-maintenance designs, none of a type which has been built before. Since there are more economical solutions which can still be designed as landmark structures, this seems a little odd.

One problem with this kind of false design contest, where the public are asked to pick from a single designer's work, is there is little opportunity for technical challenge, and also artificially constrained visual options - here, you can have any bridge you want, so long as it's white. A proper design competition would have thrown up more idiosyncratic ideas, and perhaps motivated those involved to maximise the quality of their work.

Cleveland have a survey online where you can choose your favourite of the six design options. I tried really hard to choose one, but I dislike them all pretty much equally. What do others think?


Anonymous said...

Sorry to drag this back to the Calgary competition but it looks like the vague "Arty" proposal always wins....practical concepts are never as thin, white or sculptural because they have to be built.

I think the design choices given highlight the problem engineers (as you clearly layout in your post) have with concepts, the lack of real structural design in the concept. (I have worked with Rosales early in his career and well, I will leave it there.)

I will give Rosales credit the designs are interesting.

The thing that bugs me the most is I will be accused of "sour grapes" because I question if they can be built as shown.

The lesson I continue to learn is to design the most far out thing I can and after I win, modify it back to something that can be built! Then I will be a genius! So simple I should have thought of it sooner.....

Anonymous said...

In my experience, whenever people are asked for multiple designs they'll invariable (subconsciously or not) have a favourite which will attract more of their effort and thought.

If six designers each champion one design, those design would seem much more likely to be of equal standard than if one designer champions six.

The Happy Pontist said...

There are certainly plenty of design contests which seem to be won with an unfeasible proposal which is modified later. Also several where the winning design is clearly not buildable within budget and the client later has to find extra funding.

I'm not sure the lesson is that you have to be cynical to win. An alternative is to maintain your integrity and just not enter the contests where flash is likely to prevail over value.

I've not been greatly impressed with Rosales' design work, I think he has little real sympathy with structural engineering, which makes a focus on the somewhat specialist area of bridge design difficult to understand.

The Happy Pontist said...

I think this is definitely right and was very well illustrated in Rosales' Portland scheme, where his less-preferred solutions were quite deliberately made to look less attractive than his favourite option.

Paper Project said...

The issue of bridge 'visualisations' and their sometimes tenuous connection to reality is a significant one, but the portland example you linked to really hammers home the point. Great work as always on the blog!

On the buildability of 'cynical' schemes, perhaps a one page leaflet on 'How to procure a bridge design' could be compiled by volunteers and mailed to any new bridge client:

We, the above, have seen many bridges run over time and budget. To prevent this we strongly and compassionatley recommend:

1) Employing a structural engineer on your judging panel.
2) Employing a QS on the judging panel
3) ...
4) ...

The Happy Pontist said...

A guide on how to run a bridge design competition is being prepared by IABSE at present, and once available it's hoped it will be widely publicised by RIBA, ICE and IStructE in the UK, and other bodies elsewhere. I'm not optimistic about how well it will be adopted, as most bridge design competitions are run by clients who have never used the approach before, and they rarely look to see what advice is available.